Testing Services for Replication Competent Lentiviruses

Despite current safeguards in LV production systems to eliminate the possibility of RCL generation,
there remains a low risk of them contaminating LV lots and therefore appropriate tests should be
applied to test for their presence.
The presence of RCL can be determined by several means. For example, following infection of a
susceptible cell line with RCL and serial passaging of successive cell supernatants to achieve RCL
amplification, quantitative real-time PCR can be used to detect integrated gag/pol specific nucleic
acids. For instance, for HIV-1-based LV manufacture, any RCLs generated would only share the gag
and pol genes with HIV-1. Detection of such RCLs could be based on HIV-1 Gag protein or gag gene
sequences, for which there are well-validated and sensitive assays already available, e.g. p24 Gag
immunocapture assay or gag RNA PCR assay, respectively. In addition to Gag/Pol, RCLs would
express the vector envelope protein, e.g. VSV-G. Therefore, where VSV-G is the pseudotyping
envelope protein, it is also appropriate to use a VSV-G immunoassay and/or molecular assay for VSVG
DNA or RNA to detect RCL replication. Alternatively quantitative RTase assays, following RCL
amplification involving several serial passages of a susceptible cell line, should be considered.
As a further alternative, “marker rescue” assays, in which a specific marker gene is rescued by RCL,
could also be considered for their detection.
In general, the limit of quantification required for RCL tests should be established according to the
proposed LV dose and to the size of the production lot. Ideally, the capacity of tests to detect one RCL
in a vector dose should be proven. However, the choice of a suitable, representative positive control or
reference standard is critical for demonstrating the sensitivity of RCL assays.
Any RCL assay should be well characterised and include an appropriate positive control. However, for
tests requiring cell infection the current lack of generic reference materials poses a challenge for assay
calibration. Presently, it is considered undesirable to generate a lentivirus encoding a heterologous Env
(e.g.,VSV-G) and Gag/ Pol simply to monitor RCL infectivity assays, since such a virus could be
pathogenic. For HIV-based LV, an attenuated HIV lacking all accessory genes as a positive control for
RCL assays could be of value.